MATH1041 – Lecture 1 (T1 2024) – Course Introduction & Chapter 0 June 2

Course Logistics and Roles

  • Lecturer & Convenor: Dr Yudi Bunjaman ("Yudi/Bun-ja-man", call him “Yi”)
    • Both administrative (convenor) and teaching (lecturer) duties this term
    • Email: y.bunjaman@unsw.edu.au (easy to remember, use UNSW email only)
  • Course code/title: MATH1041 – Statistics for Life & Social Sciences
    • 4 h of lecture per week, delivered in-person + live-streamed (≈12 s delay) and recorded
    • Uses long-standing common slide-deck shared by previous lecturers (Pierre, Laur, Jacob, David, Diana, …) with Yi’s custom annotations
    • Slides with blanks in class; full annotated PDF uploaded end-of-week

Communication and Support

  • Central School of Maths & Stats office email for generic admin; Yi for course-specific matters
  • Mental Health Connect: free counselling & wellbeing
  • Equitable Learning Services (ELS) for learning provisions if impediments
  • Moodle announcements = binding; check ≥ 2–3 × week
  • Moodle forum monitored by Lorraine + staff; look for existing threads before posting
  • PASS (Peer Assisted Study Sessions) available; announcement in Week 2
  • Maths Drop-in Centre re-opens Week 2 (online & face-to-face)
  • Lecturer consultation hours published Week 2 (planned adjacent to lecture slots)

Assumed Knowledge & Suitability

  • Calculus-free syllabus
    • Assumed: ≥60 in HSC Mathematics Advanced or ≥70 in Mathematics Standard (or equivalent)
    • Calculus helpful for intuition but not required
  • Not appropriate for those pursuing professional / research statistics careers – take calculus-based statistical courses instead
  • Mathematical maturity & willingness to “struggle productively” stressed; memorisation alone insufficient

Lecture Delivery & Recordings

  • 12 s online lag → ask chat to confirm audio at start
  • Public holiday in Week 2 (Mon)
    • Compensated by releasing Week 3 lecture recordings early
    • Lecture theatre sessions: finish Ch 2 (Week 1–2 content) then jump to Ch 4; parallel self-study of Ch 3 via recordings
    • Benefit: Week 3 material receives lab support within first 3 weeks of labs
  • Think–Pair–Share strategy
    • 1 min think solo → 1 min jot → 1 min discuss with neighbour → share
    • Encourages “thinking like a statistician” for life & profession
  • Regular use of Kahoot quizzes (hard online due to lag; encouraged to attend in person)

Tutorials, Labs & Online Alternatives

  • Weekly tutorial (classroom) – mostly in-person; small online class
    • Attendance expected; if miss your class:
      • Attend the online tutorial live (time listed in UNSW timetable) OR
      • Watch tutorial recording; no email needed
  • Labs (Weeks 1–3 only)
    • Purpose: gentle R + RStudio onboarding; hands-on help with Mobius lessons
    • Not compulsory but highly recommended
    • May attend multiple lab sessions if extra help needed
    • Bring USB-C/USB headphones or own laptop for video audio; captions available
    • Lab demonstrators answer any course question, not just lab sheet issues

Weekly Mobius Lessons

  • Released Tue 15:00 Week 1 → Week 11 already visible
  • Due Tue 15:00 following week
  • Formative: unlimited attempts, “How Did I Do?” button, best score kept
  • Easiest marks; use forums & consultations for help

Assessment Overview (10 UOC)

  • Weekly Mobius Lessons – 10 %
  • Lab Test 1 (Week 4) – 10 %
  • Lab Test 2 (Week 10) – 10 %
  • Assignment now split • Part 1 – due Week 5 • Part 2 – due Week 9 • Entire task released Week 3/4; feedback on Part 1 before Part 2 starts
    • Split introduced this term to distribute workload; feedback welcome
  • Final Exam – 50 %
    • Entirely computer-based (RStudio + LMS)
  • Lab Tests:
    • Practice test provided; real test ⊂ practice bank (randomised numbers/inequalities) except 1 new Q in Lab Test 1
    • Students often surprised; DO the practice!

Special Consideration

  • Apply via UNSW’s system for assessment disruption beyond your control (illness, etc.)

Study Advice & Mindset

  • Stats ≠ memorisation; requires conceptual understanding & contextual thinking
  • Hierarchical course: each topic builds on prior; clarify doubts early
  • Expect initial confusion → ask questions in lecture, tutorials, labs, forum, consults
  • Be patient with yourself; lecturer promises reciprocal patience

Course Materials & Slides

  • Tutorial Exercise Booklet: small 2024T1 update – download latest PDF
  • Textbook (OpenIntro Stats – Diez, Çetinkaya-Rundel, Barr): optional; slides are self-contained; library copies and free options
  • Pierre’s free e-book “YaRrr! The Pirate’s Guide to R” (many languages, incl. Indonesian) linked on Moodle

Software: R & RStudio

  • All assessments done in R via RStudio
    • R and RStudio Desktop free to install; university lab machines pre-configured
    • Labs give step-by-step intro; “try, error message, fix” encouraged
  • Extra R manual specific to course provided on Moodle

Chapter 0 – Why Statistics?

  • Course = introduction, scratches surface; later discipline-specific stats will build on foundations
  • Statistics: “…collecting, organising, analysing & interpreting data”
  • statistic (lower-case) = numerical summary of a dataset (mean, SD, …)
  • Everyone benefits from statistical literacy (COVID-19 dashboards cited)

Four Major Course Themes

  1. Study Design – Planning & data collection
  2. Descriptive Statistics – Summaries & visualisations
  3. Probability Theory – Mathematical foundation (recordings supplied early)
  4. Statistical Inference – Confidence intervals, hypothesis tests, etc.

Learning Outcomes (abridged)

  • Recognise when/why statistical methods are useful in your discipline
  • Choose appropriate analysis for given research problems
  • Understand & apply principles of proper study design
  • Compute & interpret CI\text{CI}s and hypothesis tests
  • Use R to perform analyses, manage data, and present results
  • Communicate statistical findings clearly

Research Questions & Data Context

  • Statistics course emphasises question-driven data collection – unlike some data-mining situations where data exist first
  • Example: “Is MATH1041 a good course?” reveals importance of clarifying:
    • What does “good” mean? (enjoyment, grades, preparation?)
    • Population (current students? past? multiple terms?)
    • Data needed (surveys, grades, comparisons, benchmarks)
    • Possible biases, measurement tools
  • Example #2: “Study a flu epidemic” – need to specify:
    • Aim (track spread? vaccine efficacy? demographic susceptibility?)
    • Population (city residents? hospital patients?)
    • Variables (infection status, date, symptoms, vaccination record, …)

Standard Steps in a Statistical Investigation

  1. Formulate research question
  2. Decide what data are required
  3. Plan how data will be collected (study design)
  4. Collect & store data appropriately (ethics, security)
  5. Describe data (tables, graphs, descriptive statistics)
  6. Analyse (model relationships, make inferences)
  7. Interpret & communicate results in context

Key Terminology (will recur all term)

  • Dataset: full collection of recorded data (often rectangular)
  • Observation / Case: single unit on which measurements taken (row)
  • Variable: recorded characteristic (column)
  • Population: entire group the study aims to understand
  • Sample: subset of population actually observed
    • Must be contained within population
  • Sample size: nn = number of observations/cases
  • Number of variables: pp (rarely used symbolically in course because pp later denotes probabilities)
  • ID/Label: non-informative identifier (ZID, random code) to distinguish cases

Illustrative Data Examples

• Course-mark file

  • Population = all MATH1041 students
  • Case = one student
  • Variables = marks for Lab 1, Lab 2, Assignment 1&2, Weekly Mobius, Final
  • Sample = all students with at least one recorded mark
  • Sample size = n=n = number of such students

• Sleep & Exercise vs Marks Study (Kahoot demo)

  • Research Q: Do average hours of sleep & physical activity predict final mark?
  • Population: all UNSW students
  • Sample: 100 randomly selected students surveyed
  • Variables (p=2p=2): weekly sleep (hours), weekly exercise (hours/minutes)
  • Sample size: n=100n=100

Course Surveys

  • Two Moodle surveys to be completed by end of Week 1
    1. Elevator/Lift behaviour (named, prize attached)
    2. MATH1041 cohort demographics & habits (anonymous)
  • Data sets will be analysed later for live examples; also inform teaching adjustments

Additional Notes & Meta-Information

  • Icons on slides: denote “extra reading”, “interactive element”, etc.; legend provided in slides PDF
  • After-class annotated slides include typed notes in place of Yi’s handwriting; Wednesday room uses Wacom for clearer ink
  • 42 students named “David Chen” on UNSW roll – illustrates why IDs > names for uniqueness
  • Humorous aside: lecturer dislikes hearing his own recorded voice – please use headphones in labs!

Ethical / Practical Implications Discussed

  • Bad statistics often arises from using incorrect analysis for question; importance of understanding vs blindly applying
  • Study design sometimes dismissed as “common sense” elsewhere, but in this course emphasised as foundation; poor design → meaningless results
  • Data security & anonymity (e.g.
    • Prize survey not anonymous, marks anonymised by ZIDs)

Wrap-Up of Lecture Session

  • Chapter 0 finished; began Chapter 1 (Study Design intro) before break
  • Next session (after 10-min break) continues Chapter 1
  • Reminder: keep confirming audio in Zoom/Teams chat; 12 s delay persists
  • Next lecture Wednesday (different theatre with improved Wacom annotations)